# Infinite Mixture of Infinite Gaussian Mixture Model

I'm struggling to implement the following model in R:

When I generate H, I get a matrix. For the DP, I am using the Chinese Restaurant Process (CRP). However, I'm not sure how to incorporate a matrix, $H$, into the CRP. I only have inputs for number of customers and my dispersion/concentration parameter, alpha. The output of my CRP is the table assignments. For example, 100 customers will be assigned to maybe 15 different tables. I know I should use these tables to weight my normal distributions, but I'm not sure what this means.

Any guidance would be helpful.